hlop3z

hlop3z/claude-ai-dangerous

Command Claude Code

Build-vs-adopt gate — for each critical concern, decide Rent/Adopt/Extend/Fork/Build and record it as an ADR.

hlop3z/claude-ai-dangerous

Command Claude Code

Generate a HANDOFF.md snapshot of session state for cross-session continuity.

hlop3z/claude-ai-dangerous

Command Claude Code

Reverse-engineer OpenSpec into a single self-contained PROJECT.md so OpenSpec can be removed without losing what matters.

hlop3z/claude-ai-dangerous

Command Claude Code

Implement tasks from an OpenSpec change (Experimental).

hlop3z/claude-ai-dangerous

Command Claude Code

Archive a completed change in the experimental workflow.

hlop3z/claude-ai-dangerous

Command Claude Code

Enter explore mode - think through ideas, investigate problems, clarify requirements.

hlop3z/claude-ai-dangerous

Command Claude Code

Propose a new change - create it and generate all artifacts in one step.

hlop3z/claude-ai-dangerous

Command Claude Code

Sync delta specs from a change to main specs.

hlop3z/claude-ai-dangerous

Settings file Claude Code

Agent settings declaring 40 allowed tools.

hlop3z/claude-ai-dangerous

MCP server Claude CodeCodexCursor

MCP server "context7" as configured in hlop3z/claude-ai-dangerous. Launched with cmd /c npx -y @upstash/context7-mcp.

hlop3z/claude-ai-dangerous

MCP server Claude CodeCodexCursor

Local-first knowledge management combining Zettelkasten with knowledge graphs. Runs locally from the basic-memory Python package.

hlop3z/claude-ai-dangerous

MCP server Claude CodeCodexCursor

MCP server "sequential-thinking" as configured in hlop3z/claude-ai-dangerous. Launched with cmd /c npx -y @modelcontextprotocol/server-sequential-thinking.

hlop3z/claude-ai-dangerous

Instructions file

Instructions for hlop3z/claude-ai-dangerous, covering how to develop here, the canon comes first, the rules (open the file when the trigger fires), the two ideas that make this work and where the rules live (don't restate them).